Mt Hutt

NZSki was rebuilding the Mt Hutt website at a time when search is everywhere: Google, AI assistants and answer engines all shape how people find a mountain. To be found and cited across all of them, GEO had to run through the whole rebuild, not get added at the end.

A site is only restructured every few years, so doing the work at any other time means fighting the architecture you just shipped. A rebuild is the cheapest moment to fix what AI engines read.

Data Story was the GEO and UX partner on the rebuild, working alongside NZSki's development team. We built the GEO work into the project: clean, canonical pages that AI engines can read and cite, then measured the result directly against the engines.

The Mt Hutt rebuild was led by customer experience, and the same project was the rare chance to fix what AI engines read. The highest-leverage move was a cleaner site, not just a better-looking one: collapse overlapping and duplicate pages into single canonical pages, set clean URLs and canonical tags, and decide deliberately what each page is for. Fewer pages, each saying one thing clearly, is what lets an engine treat the brand's own page as the source worth quoting.

Before launch, the team mapped and validated full redirects so search equity carried across the migration, set canonical and no-index rules, and built a deliberate internal-linking map so engines can parse what each page is.

The pages themselves were built to GEO best practice, so an answer engine can lift a clean passage and cite it: clear, descriptive headings, scannable sections, lists and direct answers to the questions people actually ask, and structured data describing the resort. A curated llms.txt named the source-of-truth pages for AI engines. As the copy was written, every page was run through a structured SEO and GEO editorial check, scoring how cleanly an answer engine could extract and cite it, with rewrites only where a gap was clear.

Measured directly against the major AI engines, Mt Hutt now holds the top share of voice on its core resort topics, and the engines cite the brand's own pages rather than a third party's.

Compared like-for-like on the same engines before and after the rebuild, the number of Mt Hutt pages the engines cite on those core topics held and edged up, rather than falling. For a platform migration, where the usual risk is that engines lose track of your pages entirely, holding and growing your own citation footprint is the hard-won outcome.

On broad, unbranded discovery questions, the "best New Zealand ski resorts" type of query, Mt Hutt's own pages are still cited less than the third-party lists that dominate them, and closing that gap is the next body of work. The rebuild gave the brand a strong, defensible position on the ground it owns; extending it to open discovery is where the next phase goes.

The structure AI engines read is set during a rebuild, so building GEO into the Mt Hutt project was the cheapest, highest-leverage time to do it. Mt Hutt came through the migration found and cited across Google and the AI engines, on its own pages, with the open-discovery frontier mapped as the next step.